Vector computing has nothing to do with std::vector. It's a way of performing one instruction on several units of data. This kind of computing is used with technologies: MMX, SSE, AVX, Cell CPU (PlayStation 3).
You can achieve this kind of computing using compiler intrinsics or by exploiting compiler auto-vectorization (compiler converts your program silently to use vector instructions).
- GCC: Using Vector Instructions through Built-in Functions
- GCC: An Introduction to GCC Compiler Intrinsics in Vector Processing - George Koharchik & Kathy Jones, 2012
- Intel C/C++: How to Use Intrinsics - Joseph D. Wieber, Jr and Gary M. Zoppetti, 2008
- Intel: Introduction to Intel® Advanced Vector Extensions - Chris Lomont, 2008
- Getting started with SSE programming - The Supercomputing Blog, 2009
- How to Write Fast Numerical Code - Markus Püschel, 2011
- Programming Guidelines for Vectorizing C/C++ Compilers - Aart Bik, Milind Girkar, Paul Grey, and Xinmin Tian, 2003
- GCC: Auto-vectorization with gcc 4.7 - Lockless Inc.
- Auto-Vectorizer in Visual Studio 2012 - Jim Hogg, 2012
- x86 Open64 Compiler Suite
- Intel: A Guide to Auto-vectorization with Intel® C++ Compilers
- Vectorising code to take advantage of modern CPUs (AVX and SSE) - Walking Randomly